Update app.py

app.py

@@ -25,9 +25,9 @@ def main_interface():
                 n_layers = gr.Slider(1, 5, step=1, label="# Hidden Layers", value=2)
                 epochs = gr.Slider(10, 300, label="Epochs", value=100)
                 learning_rate = gr.Slider(1e-5, 0.01, label="Learning Rate", value=0.001)
-                beta1 = gr.Slider(0.8, 0.95, label="AdamW Beta1", value=0.9, step=0.01)  # Added
-                beta2 = gr.Slider(0.9, 0.999, label="AdamW Beta2", value=0.999, step=0.001)  # Added
-                weight_decay = gr.Slider(0.0, 0.1, label="Weight Decay", value=0.01, step=0.001)  # Added
+                beta1 = gr.Slider(0.8, 0.95, label="AdamW Beta1", value=0.9, step=0.01)
+                beta2 = gr.Slider(0.9, 0.999, label="AdamW Beta2", value=0.999, step=0.001)
+                weight_decay = gr.Slider(0.0, 0.1, label="Weight Decay", value=0.01, step=0.001)
                 dropout = gr.Slider(0.0, 0.3, label="Drop Out", value=0.2)
                 window_size = gr.Slider(5, 90, label="Window Size", value=30)
                 test_split = gr.Slider(0.05, 0.5, label="Test Split", value=0.2)
@@ -37,10 +37,10 @@ def main_interface():
 
             with gr.Column(scale=2):
                 backtest_plot = gr.Plot(label="📊 Backtesting: Actual vs Forecast")
-                future_plot = gr.Plot(label="🔮 Future Forecast")
+                future_plot = gr.Plot(label="🔮 Future Forecast with Actuals")
                 future_table = gr.Dataframe(label="📋 Future Predictions")
-                r2_plot = gr.Plot(label="📉 R² and MAPE Metrics")  # Updated
-                error_plot = gr.Plot(label="📉 RMSE and MAE Metrics")  # Updated
+                r2_plot = gr.Plot(label="📉 R² and MAPE Metrics")
+                error_plot = gr.Plot(label="📉 RMSE and MAE Metrics")
                 loss_plot = gr.Plot(label="📈 Training Loss Curve")
 
         def run_pipeline(data_src, ticker, file_upload, start_date, end_date, horizon, model,
@@ -51,41 +51,46 @@ def main_interface():
                 pd.to_datetime(end_date)
 
                 source_key = "csv" if data_src == "Upload CSV" else "yahoo"
-                df = load_data(data_src=source_key, ticker=ticker, file_upload=file_upload, start=start_date, end=end_date)
-                if df is None or df.empty:
+                main_df, future_df = load_data(data_src=source_key, ticker=ticker, file_upload=file_upload, 
+                                             start=start_date, end=end_date, horizon=horizon)
+                if main_df is None or main_df.empty:
                     return None, None, None, None, None, None, "❌ Failed to load data. Please check input."
 
                 result = get_model(
-                    df=df,
+                    df=main_df,
+                    future_df=future_df,  # Pass future_df
                     model_name=model,
                     horizon=horizon,
                     hidden_units=hidden_units,
                     n_layers=n_layers,
                     epochs=epochs,
                     learning_rate=learning_rate,
-                    beta1=beta1,  # Added
-                    beta2=beta2,  # Added
-                    weight_decay=weight_decay,  # Added
+                    beta1=beta1,
+                    beta2=beta2,
+                    weight_decay=weight_decay,
                     dropout=dropout,
                     window_size=window_size,
                     test_split=test_split
                 )
                 forecast_plot = plot_forecast(result)
-                future_plot = plot_future_forecast(df, result)
-                r2_plot = plot_metrics_r2(result)  # Updated
-                error_plot = plot_metrics_errors(result)  # Updated
+                future_plot = plot_future_forecast(main_df, result, future_df)  # Pass future_df
+                r2_plot = plot_metrics_r2(result)
+                error_plot = plot_metrics_errors(result)
                 loss_plot = plot_loss_curve(result)
 
                 msg = "✅ Done."
                 if "latest_prediction" in result:
-                    last_date = df['Date'].iloc[-1]
+                    last_date = main_df['Date'].iloc[-1]
                     future_dates = pd.date_range(start=last_date + pd.Timedelta(days=1), periods=horizon, freq='B')
-                    future_df = pd.DataFrame({'Date': future_dates, 'Predicted Value': result["latest_prediction"]})
+                    future_data = {'Date': future_dates, 'Predicted Value': result["latest_prediction"]}
+                    if not future_df.empty and "future_actuals" in result:
+                        future_data['Actual Value'] = result["future_actuals"] + [None] * (horizon - len(result["future_actuals"]))
+                    future_df_out = pd.DataFrame(future_data)
                     msg += f" Next predicted value(s): {[f'{val:.2f}' for val in result['latest_prediction']]}"
                 else:
-                    future_df = pd.DataFrame()
+                    future_df_out = pd.DataFrame()
 
-                return forecast_plot, future_plot, future_df, r2_plot, error_plot, loss_plot, msg
+                return forecast_plot, future_plot, future_df_out, r2_plot, error_plot, loss_plot, msg
             except Exception as e:
                 return None, None, None, None, None, None, f"❌ Error: {str(e)}"